Saved in:
Bibliographic Details
Main Authors: Gao, Qinjiao, Wang, Zuowei, Zhang, Ran, Wang, Dongjiang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.03244
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917976537563136
author Gao, Qinjiao
Wang, Zuowei
Zhang, Ran
Wang, Dongjiang
author_facet Gao, Qinjiao
Wang, Zuowei
Zhang, Ran
Wang, Dongjiang
contents In this paper, we propose the Adaptive Movement Sampling Physics-Informed Residual Network (AM-PIRN) to address challenges in solving nonlinear option pricing PDE models, where solutions often exhibit significant curvature or shock waves over time. The AM-PIRN architecture is designed to concurrently minimize PDE residuals and achieve high-fidelity option price approximations by dynamically redistributing training points based on evolving PDE residuals, while maintaining a fixed total number of points. To enhance stability and training efficiency, we integrate a ResNet backbone, replacing conventional fully connected neural networks used in Physics-Informed Neural Networks (PINNs). Numerical experiments across nonlinear option pricing models demonstrate that AM-PIRN outperforms PINN, RAM-PINN, and WAM-PINN in both resolving PDE constraints and accurately estimating option prices. The method's advantages are particularly pronounced in complex or multi-dimensional models, where its adaptive sampling and robust architecture effectively mitigate challenges posed by sharp gradients and high nonlinearity.
format Preprint
id arxiv_https___arxiv_org_abs_2504_03244
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Movement Sampling Physics-Informed Residual Network (AM-PIRN) for Solving Nonlinear Option Pricing models
Gao, Qinjiao
Wang, Zuowei
Zhang, Ran
Wang, Dongjiang
Networking and Internet Architecture
In this paper, we propose the Adaptive Movement Sampling Physics-Informed Residual Network (AM-PIRN) to address challenges in solving nonlinear option pricing PDE models, where solutions often exhibit significant curvature or shock waves over time. The AM-PIRN architecture is designed to concurrently minimize PDE residuals and achieve high-fidelity option price approximations by dynamically redistributing training points based on evolving PDE residuals, while maintaining a fixed total number of points. To enhance stability and training efficiency, we integrate a ResNet backbone, replacing conventional fully connected neural networks used in Physics-Informed Neural Networks (PINNs). Numerical experiments across nonlinear option pricing models demonstrate that AM-PIRN outperforms PINN, RAM-PINN, and WAM-PINN in both resolving PDE constraints and accurately estimating option prices. The method's advantages are particularly pronounced in complex or multi-dimensional models, where its adaptive sampling and robust architecture effectively mitigate challenges posed by sharp gradients and high nonlinearity.
title Adaptive Movement Sampling Physics-Informed Residual Network (AM-PIRN) for Solving Nonlinear Option Pricing models
topic Networking and Internet Architecture
url https://arxiv.org/abs/2504.03244